Entries in hypnagogia (3)


Sleep Onset Apnoeas / "Throat Closing"

One of the search terms that people find this blog with surprised me by how frequently people search for it, there are several variations but these two sum things up well:

  • Throat closes as I go to sleep
  • Choke as I fall asleep

I hadn't really considered this before but I really should have done because it happens to me, and one of my parents. Normally when I experience those throat-closing moments they are accompanied by the first snore / cluck / snort of the night, a flash of whatever broken dream images were going through my mind and the knowledge that I'll soon be asleep. I guess it's something that a lot of us experience but don't really talk about it because it sounds a little strange.

Polite and usual "sleep chat" is usually along the lines of "Did you sleep well?" and not much deeper than that which is a shame because it's such a fascinating subject that few of us speak about because many (wrongly) consider sleeping to be a waste of 8 hours and some even see it as a weakness ("Sleep is for wimps" etc etc).

I've been using CPAP for a month or so now and have been adjusting my own pressures based on the sleep study data that I record each night. When I reviewed each night's data, one of the things that stood out about breathing as we drop off to sleep was how it changes at sleep onset (the moment that a sleep lab would declare that we are now asleep).

This becomes more pronounced with CPAP (or at least for me as a relative newcomer to using CPAP for myself) because when I first switch the machine on at night I instantly become aware of my breathing and try to control it. Oddly, consciously controlling our breathing is something that we don't do very well and it is best left when our bodies do it for us automatically. Can you imagine the chaos that would result if we had to consciously take every breath or control every heartbeat?

The following is a few minutes of data representing my breathing as I was dropping off to sleep whilst wearing CPAP. It's taken from Resmed's ResScan software (click for a larger version).

The breathing on the left is when I was awake, it is clearly different to the smooth regular breaths on the right side of the graph. Falling asleep is what brought about the change as I "forgot" to consciously breathe and my body took over.

So far so good, but on several nights I noticed that the handover from from awake breathing to asleep breathing wasn't so smooth... 

Notice how there are gaps in my breathing? This is from a night where some apnoeas were still occuring; I remember that as I was falling asleep I experienced at least two of those "throat closing" moments. The apnoea briefly woke me as left me realising that a moment before I was asleep, which also seemed to shift the hypnagogia into my conscious mind and then into my memory because this night was also a night that I remembered a lot of it.

Several different images and phrases came and went as I drifted off to sleep. The Zeo recorded that I briefly entered REM as I fell asleep (possibly causing a loss in muscle tone resulting in apnoeas)...

This is something that I see on many nights when I look at my Zeo data, but in the interest of accuracy, Zeo do point out that if your sleep is generally healthy with 7-9 hours per night and no feelings of tiredness during the day that this brief period of REM could be wake being misinterpreted as REM. It could be that N1 sleep mixed with brief periods of wake (from my respiratory arousals) were interpreted as REM by the Zeo (as wake is so similar to REM). 

The following night my sleep looked very different and I achieved a lot more REM along with no apnoeas or hypopneas recorded at sleep onset, intriguingly the Zeo data doesn't show me passing through REM as I fell asleep. So this could actually be REM, although this could also be because there were no respiratory related arousals during the transition from wake to sleep.

With the Black Shadow Sleep Monitor I've seen that less significant respiratory events are linked with altered REM sleep, in my case a series of hypopneas leading to a Lucid Dream. This raises an interesting issue that may provide some insight into dream formation.

I used to be comfortable with those throat-closures before I knew I had Sleep Apnoea as they had become so familiar to me. I used to use them as a way of knowing that I'd soon be asleep. In a strange way I will be sad if they go because after each "cluck" or snort I'd briefly wake and commit the partial dream or hypnagogic image to memory and would be able to study sleep as I drifted off.

Famously people have used a technique that relies on waking from hypnagogia in order to remember it which involves napping in a chair whilst holding a metal spoon over a metal tray or plate. Once you drop off to sleep you automatically release the spoon, causing a clattering sound which then wakes you allowing you to recall what you saw and heard.

I see my sleep onset apnoeas as such a system, alebit a naturally occuring one. I think that during the times that I want the best of both worlds of having the apnoeas at sleep onset but sleeping safely for the remainder of the night that I will experiment with setting a RAMP on my CPAP which delays the maximum pressure by up to 45 minutes, giving me time to explore the hypnagogic world.

Knowing that I have obstructive sleep apnoea and that the first apnoeas of the night are usually at sleep onset, I would suggest that anyone who experiences these throat-closures at sleep onset should at the very least be aware that they could have Obstructive Sleep Apnoea and look out for other symptoms. The best thing to do it get it checked professionally as it could also be a sign of another condition such as Acid Reflux (GERD) or Laryngospasm.



You're tired and have just got into bed...

You close your eyes and stare at the blackness. Within a few seconds you see random sparks and colour flashes.

You've never really thought about them but if you have you've probably assumed that they're just the random firings of your tired optic nerves as they relax after a hard day.

It looks as if you're drifting through space floating past random galaxies and nebulae.

Then there's a bright flash to the bottom left of your vision, it whizzes off, it must have been a car driving past so, it becomes a car whizzing past on a dark road. Then there's another one. The lights are blurred, maybe it's raining and these are reflections of headlights on the wet road.

Why are you standing beside a road in the rain? Maybe your car has broken down and you are waiting for help. You sense someone behind you. Is it your passenger? Is it a stranger? Are you in trouble?

Fairly frequently, for me, this can evolve into a dream. Although many times these are just fleeting images on a black screen; sometimes they move (like the red blob that became an egg, which then hatched before vanishing) and sometimes they are just static images such as faces or objects. Most of the time the colours are exaggerated and over-saturated.

So what's going on?

My brain is doing what brains do best: it makes sense of the information that it is given. Sometimes it joins the information together to make a narrative, a story. More often than not, in a dream, it does this seamlessly so that you (as the observer, creator and participant of the dream) don't stop to think "Why is there a car driving past in space? Why am I now beside a road?", you just accept it as normal and carry on enjoying the story.

Sometimes as I watch the flashes and sparkles I am pretty much awake, certainly conscious. I am able to question those things and soon realise that the only way that I could have been in one place then magically transported to another place is if I am dreaming.

That's when sometimes I can separate out the creator / observer / participant aspects and consciously create the dream. This is called Lucid Dreaming.

I find that if I concentrate on the images as I fall asleep then as soon as I become conscious of them becoming a story that I wake up, almost as if that kick of consciousness drags me back to the waking world.

Or maybe this has happened to you...

You start to drift off to sleep when all of a sudden you hear your doorbell ring.

You jump up to answer the door only to find that there's no one there. So you go back to bed.

No sooner have you start to doze again than you hear another ring of the doorbell, but this time something is different - it's not YOUR doorbell, it sounds different. You sit up startled, and look around the room before getting back into bed.

Eyes closed, and then the phone rings once and stops. Tired, irritated, and a little scared that someone or something is playing tricks on you, you return to sleep.

Doorbells, phones, knocking, people calling your name, even short tunes wake me up, none of which are real. That is to say that none are generated by external stimuli, they are in fact hypnagogic hallucinations, and to the brain they are very real.

To be more precise, these images and hallucinations can be either hypnogogic or hypnopompic, but they are generally lumped together under the banner of hypnaogic.

Hypnogogic = going into sleep
Hypnopompic = waking from sleep

This is a well documented phenomena, and is nicely illustrated in the book "Head Trip - a Fantastic Romp Through 24 Hours in the Life of Your Brain". The whole book is illustrated in a similar fashion and is well worth a read. Original here: http://www.jeffwarren.org/illustrations/mavromatis’-four-stages-of-hypnagogia

So this is normal. This comes as a relief to me because I first became aware of these flashes and images in my late teens. I'd just started work which involved being on-call, hence my sleep was frequently disturbed and I found it difficult to sleep at night as I was forever listening out for my pager in case I was called out.

I found myself drifting off to sleep at any opportunity during the day, at a desk, in a chair etc. Then when I did sleep at night I had increasing spells of sleep paralysis, so I went to see a doctor and asked if there was anything that could be done to help, such as tablets to help me sleep when I wasn't on-call.

He asked me to describe what went through my head as I tried to sleep at night. When I mentioned the colour flashes he didn't really react, but when I said that I was always tired he decided that I must have "mild depression", he explained that it "wasn't clinical, but it was just being run-down"... and prescribed trycyclic antidepressants.

I clearly remember thinking that this was odd. I didn't feel depressed, just tired. The last thing I said to the doctor before taking the prescription was, "So these colour flashes are because I'm depressed, and these pills will stop them?". To which he replied, "Yes".

The antidepressants made me tired, which did help me sleep at night, but they also gave me disturbed dreams and nightmares and left me feeling even more tired in the day. Not a great solution, and one that I was quick to give up on.

That was nearly 20 years ago, and thankfully awareness of sleep medicine has moved on since then. Nowadays I would hope that most general practitioners would be aware of sleep paralysis and the effect of a disturbed sleep routine on general wellbeing. However, I doubt that many would be aware of the subtleties of Hypnagogia and the normal blurring of states of consciousness that every sleeper goes through on the journey to sleep - despite travelling through them themselves!

As I write this it's occurred to me that the hypnagogic sounds that I normally hear are all things that demand my attention: Phones, doorbell, children calling, explosion outside etc. Only once has it been a few notes of music that didn't cause me to sit bolt-upright in bed.

Maybe this is a hang-up from the being on call nights or having to wake at the drop of a hat to deal with my son's prolonged periods of apnea for many years?

Could it be that over the years a brain learns to be in a "hypersensitive" state, ready to wake up and so is fooled by dream sounds? If so, why would this only happen in REM. This may be similar to Wehr's experiment showing that if a subject expected to be disturbed then the hormone Prolactin wouldn't be released during sleep.

"Wehr quickly discovered that prolactin was vulnerable to almost any disturbance. Simply talking to the subjects would interrupt its secretion, as would their expectation that someone was going to talk to them. It was a fragile state: the subject had to be lying in the dark, expecting not to be disturbed, for the drug to work. But when it did work, it appeared to produce a period of gentle quiescence, a pleasant, meditative state in which time passed very quickly for the subjects.

In addition, each period of quiet rest, wrote Wehr, always emerged directly from “particularly intense” periods of rem sleep featuring vivid dreams, full of emotional resonance."

(Jeff Warren, "Head Trip, a Fantastic Romp Through 24 Hours in the Life of Your Brain)

Hormones aside, I know that certain things that I do can induce these sounds. One example is if I sleep with a fan on in the room. I suspect that this is because the fan generates "white noise" which makes it harder to hear if I am really called in the night, so my brain "plays safe" by alerting me to the dream noises that, if they were real, would demand my attention.

Hypnagogic hallucinations can be actually involve all the senses...

Someone saw the painting of Fuseli's "The Nightmare" on one of my walls and asked me a question. They asked, "Do you ever get walked on at night?". Needless to say this threw me and I wasn't quite sure where they were taking the conversation, but they continued.

They explained that they had been on a holiday in an old stone cottage recently and that during the night they felt as if someone had walked through the room, across the bed and up the stairs in the bedroom.

Already I suspected that this may have been a parasomnia and asked them to tell me more about the day.

I won't recant the whole tale here, but the beginning was enough to confirm that it was a likely hypnagogic episode or even a form of sleep paralysis.

"It'd been a long drive and it was late when I arrived. I made the bed up and got straight it. Just as I was dozing off I felt the bed crumple as if it'd been walked on. I laid there as I heard the footsteps walking up the stairs"

There are several pointers here:

  • Strange room
  • Tired
  • Presence in the room

...and lastly (which is where this blog began)... 

  • Tired and just got into bed

I'll go into more detail about this type of event when I blog about Sleep Paralysis.

I've got a particular fondness for hypnagogia (thanks to the episode with the doctor) so I tried to capture what is going on when it occurs. 

I don't have a full EEG to experiment with, but I do have the raw output from the Zeo's single channel EEG. Although this only looks at the front of the brain (combined with muscle tone and eye movements), it still gives a good guide to what is happening when these episodes occur.

I've concentrated on the episodes that take place in the middle of the night.

When a phantom doorbell rings (or whatever form it takes) then I look at the clock. More often than not I remember the time in the morning and check what the Zeo picked up at that time.

From this data I have learned that most of the hypnaggogic hallucinations occur when I am flitting between sleep stages.

Click for a larger image

The 5 minute eopch of the whole Zeo hypnogram (not shown above) showed me as being awake for this period, which is unsurprising as the live raw data and the 30 second hypnogram showed me as flitting between wake and N1/N2 (and having come from N2 - indicated by sleep spindles).

What isn't clear to me is whether these noises are the cause or effect of sleep state transition. ie, did I wake up then re-doze, dragging some lingering elements of consciousness to the sleeping world with, or did the noise (and presumable unremembered noises cause me to wake up)?

Sadly, without a more precise time that the event occurred and a more detailed EEG I doubt that I will be able find an answer, or be able to point at a wiggly line and say, "THERE is the noise that woke me".

What I can see from looking at the raw data is that the event occurred in the minute of 2322h and I stirred for a moment before looking at the time, resulting in w1 (wake 1 shown by a red peak representing delta waves along with the corresponding mass of noisy delta waves on the EEG line) at 2323:04h.

Interestingly, there is a spike (d1) of delta waves which are much neater in appearance, much like those of slow wave (delta) sleep a few minutes beforehand. This was interpreted as an awakening, but I suspect that it wasn't (it looks too noise-free). 

What I'll have to hope for is a more prolonged episode to look at, or perhaps being lucky enough to catch an episode of sleep paralysis whilst I'm recording the raw data.



5-HTP, Serotonin and Sleep



For the last 5 days I've been taking 5-HTP.


5-HTP is a precursor to serotonin (that is, it becomes serotonin once metabolised by the body).


I've previously used 5-HTP for experimenting with my sleep composition, and to try to reliably trigger lucid dreams. The dosage that I was taking then was 50 - 100mg at night.


This time I tried a brand that contained "co-factors". These are associated vitamins and minerals that help the body to metabolise the 5-HTP into serotonin. 

I decided on 200mg each night about 10 minutes before getting into bed. 100mg is the recommended maximum daily dose of this brand, although I have seen others that give a maximum daily dose as 300mg.

I have also seen medical literature that cites doses of 150mg-300mg as a daily dose (as a trial for treating depression), so toxicity at this dose seemed unlikely.

Why did I take 5-HTP?

The rather lazy answer is that (as I mentioned above) I've taken it before without ill effect, and that it is mentioned to have a positive effect on sleep, even being cited as having a positive effect on a severe case of insomnia 

A rather cheeky and over-simplified answer is that I have already shown that a depressant (alcohol) increases my AHI, so I wondered if something that has antidepressant properties would have the opposite effect and reduce my AHI.

I have also seen medical literature that suggests a link between depression and sleep apnoea (both as a cause and effect of sleep apnoea). 

Depression is more complicated that just having low serotonin levels, but low serotonin does play a part. Serotonin is a neurotransmitter (it is used in the body's sending and receiving of nerve impulses), so I wondered if increasing my levels would mean that I would have stronger signals to breathe during sleep and that the muscle tone in my airways would be that much "sharper".

Okay, so how did the 5 days go?

I noticed the following effects (sleep related and non-sleep related):

  • I wasn't as tired in the evenings, so went to bed later
  • I found it easier to wake up
  • I didn't feel as hungry during the day, generally ate once at 2pm, then didn't eat in the evening.

Throughout these experiments, I go to bed when I am tired.

I calculated my average bedtime using the Zeo data. As I am doing these 5-day comparisons on weekdays (except the alcohol test - see separate blog for that), the time that I had to wake up was the same each day, hence later to bed meant less time in bed, which oddly with the 5-HTP didn't feel like a bad thing.








I wondered if my body "needed" more sleep, afterall I am used to roughly an extra hour (and still crave more), but on two mornings of the week (Fri and Sat) I can have a bit of a lie-in if I need to but I didn't want to lie in when taking 5-HTP at night (It's Saturday and I'm typing this at 7am, after waking up at my normal weekday time of 6:30am)!

It is known that some anti-depressants can decrease the amount of REM sleep that you have, so I suspected that my percentage of REM would decrease.

As percentages, the figures do not show this. In fact they show an increase in the percentage of deep sleep, along with a slight increase in REM and wake...

However, as REM is more abundant at the end of the night, it follows that the less time one spends in bed, the less opportunity there is for REM sleep, hence my actual 5-day-mean time in REM fell by 11 minutes from my baseline of 98 minutes, although it put it more inline to the average for my age of 90 minutes.

My actual time in slow-wave sleep (deep) rose by a mean of 3 minutes.  My mean deep sleep (34 minutes) is still way below the average for my age (69 minutes) but within limits, so I am pleased with this small increase.

I'm not sure if these changes are actually significant, perhaps a longer trial may reveal more?

So, even if 5-HTP didn't affect my AHI, it still had positive effects on several aspects of my sleep composition. 

However, it did have a positive effect on my AHI and it was much more noticeable... 

I've graphed my 5-day-mean AHI alongside the other 5-day-mean AHIs for alcohol and my baseline.


It brings my mean AHI down to a level that puts me in the "normal" category (<5 AHI = normal). Maybe this is the reason for not wanting to sleep so early and finding it easier to wake in the mornings.

For the sake of showing that this is a consistent effect, I also graphed the data on a night by night basis... 

For now, this seems too good to be true: 

  • Lower AHI
  • Normal bedtimes
  • Less time in bed
  • Increase in slow-wave sleep

So what now? Keep on taking the tablets?

Based on my limited data, I wouldn't dare go so far as to claim that this is a effective treatment for mild (very mild) obstructive sleep apnoea, nor would I suggest that it would work for others, but it is intriguing and does need looking at further.

It leaves me wanting to know more about the mechanism behind this effect. Yes, it's good that it has helped me as far as my AHI is concerned, but I want to know how. In a previous post, I noted that my sleep apnoea seems to be REM related. Serotonin related activity drops dramaticaly during REM sleep, so maybe the higher levels due to 5-HTP reduced that effect?

Is my pseudo-science hypothesis correct, or is there more to it than that?

I'd be interested to see if this effect carried on, or if my body got used to the 5-HTP and the effect faded.

I'd also like to try this with a lower dose of 5-HTP (after all, why take a high dose if you don't need to)? I'd also like to see if sustained-release 5-HTP is more effective.

I was planning to try the mandibular advancement device next, but after such a positive effect, I think I'll stay with 5-HTP for a while and see if I can improve on the results even more. I'd like to bring my sleep latency (time to sleep onset) down.

Yes, this is good news but I can't help feeling a little like Lizzy in "Drop Dead Fred", she knows that she's taking a pill that will stop her seeing things that others can't. Will taking 5-HTP, increasing my serotonin levels and bringing my AHI down stop me from having sleep paralysis, lucid dreams and seeing/hearing the sleep-wake border imagery that I have grown so used to? 

I hope not.

I'll explain how and why I came to like sleep paralysis in an upcoming blog-post. Hopefully it may be of use to anyone that fears it as I used to.


I have to say that this is a test with a tiny sample size, and of limited duration. I am not suggesting that anybody should try this, and certainly not use it in place of recognised treatments.